Technology companies recognize that continuous innovation is needed to keep pace with competition. It’s why Amazon bumped product development investment by 28% in 2016 and Facebook increased innovation and product development dollars in 2016 by 26%.

These are big numbers, so it comes as no surprise that many companies use a scientific metrology like predictive analytics to help determine product direction and how they’re going to spend their innovation dollars.

Netflix uses analytics to classify the key attributes of past and current products or services and then models the relationship between those attributes and the commercial success of the offerings to derive guidance for new product undertakings–and Proctor & Gamble uses predictive analytics to predict the likelihood of market success.

But there’s also the contrarian approach of “gut feel,” where instinct as the purest form of intelligence takes center stage in product development.

To illustrate the difference between predictive analytics and gut feel approaches to new products, here are two very different use cases from the food service industry:

McDonald’s and predictive analytics

McDonald’s made the move to digital menus in its restaurants, in part because it enables the restaurant chain to instantaneously modify menus to product test new menu item concepts in certain regions of the country. Based upon predictive analytics, McDonald’s restaurants can alter these digital menus based upon the time of day, or even the weather, in order to tailor menu offerings to what customers are most likely to order. The fast food chain said that the combination of analytics and malleable digital menus has increased revenues and facilitated more new product proofs of concept

Seattle Mariners and gut feel

“Seattle is an awesome food scene, so every year when we decide how we are going to tweak the food and beverage items that we offer at SafeCo Field, we visit local restaurants and sample the menus for ideas,” said Steve Dominguez, General Manager of the Centerplate concession program at Safeco Field for the Mariners.

This year, Dominguez came across roasted grasshoppers, a Mexican favorite that a local restaurant was serving. “The roasted grasshoppers was a different idea and we took a collaborative approach to it,” said Dominguez. “We invited team management to a sampling of the grasshoppers and asked them what they thought. The consensus was, Let’s do it!”

So with a gut feel hunch and no analytics, the Mariners offered roasted grasshoppers on opening day. Fans consumed 18,000 grasshoppers and the concession sold out. The grasshoppers were so hot that orders actually had to be limited, and Dominguez was soon overnight-shipping grasshoppers to the park because local supplies could not meet demand.

“Each year, we try to reinvent the menu to give it more splash,” said Dominguez. “Guys from other ball clubs were calling about the grasshoppers and asking, ‘What are you guys doing up there?’ We didn’t use analytics–we just thought it would work.”

So if both analytics and gut feel approaches to new products can work, is there a middle ground that can take advantage of both?

Yes–there is a middle ground between going 100% analytics, where you risk missing out on great ideas just because your analytics can’t confirm them; and 100% gut feel, where you don’t have any analytics to give you at least some confirmation for your idea before you try it.

Here are three recommendations for companies that want to incorporate both gut feel and analytics in new product development:

Create a product development intake process that can accommodate both gut feel ideas and the ideas that come out of your analytics.

You can do this by making a list of candidate new product proposals that come in either through the gut feel or the analytics door. This might seem straightforward, but for many organizations that are now strictly focusing on what their analytics tell them because of a fear of being wrong, it isn’t. Instead, the goal should be to keep all doors to new ideas open, so you don’t leave any creativity on the table.

Secure the support of your management

More conservative businesses (e.g., insurance or finance) are less likely to invest in gut feel product development, because they are analytic by nature and they feel more comfortable that their R&D dollars will be spent wisely if they have analytics to back up an idea. Conversely, other industries (e.g., technology or entertainment) leave more room for gut feel ideas. Before you initiate a dual analytics-gut feel approach to your product innovation, understand what your corporate culture is likely to support. If you can’t get upper management’s backing for a dual approach, the concept will not work.

Whether you are using predictive analytics or gut feel, know when to pull the plug

The most well conceived analytics study or the best feel-good idea isn’t always going to work. R&D is innovation, and innovation includes its share of failures as well as successes. As soon as an idea beings to show that it will not succeed, pull the plug. In this, way you manage your losses and control your risks.

Also see:
Predictive analytics are not a crystal ball: What CXOs can learn from the 2016 election
Infographic: 7 ways to build trust in data and analytics at your company
Risks and rewards of making project decisions based on gut feelings
Analytics prediction confidence: It’s all about semantics